]>
Commit | Line | Data |
---|---|---|
1 | import itertools | |
2 | import json | |
3 | import random | |
4 | import re | |
5 | import string | |
6 | import time | |
7 | ||
8 | from .common import InfoExtractor | |
9 | from ..compat import compat_urllib_parse_unquote, compat_urllib_parse_urlparse | |
10 | from ..networking import HEADRequest | |
11 | from ..utils import ( | |
12 | ExtractorError, | |
13 | LazyList, | |
14 | UnsupportedError, | |
15 | UserNotLive, | |
16 | determine_ext, | |
17 | format_field, | |
18 | get_element_by_id, | |
19 | get_first, | |
20 | int_or_none, | |
21 | join_nonempty, | |
22 | merge_dicts, | |
23 | qualities, | |
24 | remove_start, | |
25 | srt_subtitles_timecode, | |
26 | str_or_none, | |
27 | traverse_obj, | |
28 | try_call, | |
29 | try_get, | |
30 | url_or_none, | |
31 | ) | |
32 | ||
33 | ||
34 | class TikTokBaseIE(InfoExtractor): | |
35 | _APP_VERSIONS = [('26.1.3', '260103'), ('26.1.2', '260102'), ('26.1.1', '260101'), ('25.6.2', '250602')] | |
36 | _WORKING_APP_VERSION = None | |
37 | _APP_NAME = 'trill' | |
38 | _AID = 1180 | |
39 | _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s' | |
40 | _WEBPAGE_HOST = 'https://www.tiktok.com/' | |
41 | QUALITIES = ('360p', '540p', '720p', '1080p') | |
42 | ||
43 | @property | |
44 | def _API_HOSTNAME(self): | |
45 | return self._configuration_arg( | |
46 | 'api_hostname', ['api16-normal-c-useast1a.tiktokv.com'], ie_key=TikTokIE)[0] | |
47 | ||
48 | @staticmethod | |
49 | def _create_url(user_id, video_id): | |
50 | return f'https://www.tiktok.com/@{user_id or "_"}/video/{video_id}' | |
51 | ||
52 | def _get_sigi_state(self, webpage, display_id): | |
53 | return self._parse_json(get_element_by_id( | |
54 | 'SIGI_STATE|sigi-persisted-data', webpage, escape_value=False), display_id) | |
55 | ||
56 | def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True, | |
57 | note='Downloading API JSON', errnote='Unable to download API page'): | |
58 | self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choices('0123456789abcdef', k=160))) | |
59 | webpage_cookies = self._get_cookies(self._WEBPAGE_HOST) | |
60 | if webpage_cookies.get('sid_tt'): | |
61 | self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value) | |
62 | return self._download_json( | |
63 | 'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id, | |
64 | fatal=fatal, note=note, errnote=errnote, headers={ | |
65 | 'User-Agent': f'com.ss.android.ugc.{self._APP_NAME}/{manifest_app_version} (Linux; U; Android 13; en_US; Pixel 7; Build/TD1A.220804.031; Cronet/58.0.2991.0)', | |
66 | 'Accept': 'application/json', | |
67 | }, query=query) | |
68 | ||
69 | def _build_api_query(self, query, app_version, manifest_app_version): | |
70 | return { | |
71 | **query, | |
72 | 'version_name': app_version, | |
73 | 'version_code': manifest_app_version, | |
74 | 'build_number': app_version, | |
75 | 'manifest_version_code': manifest_app_version, | |
76 | 'update_version_code': manifest_app_version, | |
77 | 'openudid': ''.join(random.choices('0123456789abcdef', k=16)), | |
78 | 'uuid': ''.join(random.choices(string.digits, k=16)), | |
79 | '_rticket': int(time.time() * 1000), | |
80 | 'ts': int(time.time()), | |
81 | 'device_brand': 'Google', | |
82 | 'device_type': 'Pixel 7', | |
83 | 'device_platform': 'android', | |
84 | 'resolution': '1080*2400', | |
85 | 'dpi': 420, | |
86 | 'os_version': '13', | |
87 | 'os_api': '29', | |
88 | 'carrier_region': 'US', | |
89 | 'sys_region': 'US', | |
90 | 'region': 'US', | |
91 | 'app_name': self._APP_NAME, | |
92 | 'app_language': 'en', | |
93 | 'language': 'en', | |
94 | 'timezone_name': 'America/New_York', | |
95 | 'timezone_offset': '-14400', | |
96 | 'channel': 'googleplay', | |
97 | 'ac': 'wifi', | |
98 | 'mcc_mnc': '310260', | |
99 | 'is_my_cn': 0, | |
100 | 'aid': self._AID, | |
101 | 'ssmix': 'a', | |
102 | 'as': 'a1qwert123', | |
103 | 'cp': 'cbfhckdckkde1', | |
104 | } | |
105 | ||
106 | def _call_api(self, ep, query, video_id, fatal=True, | |
107 | note='Downloading API JSON', errnote='Unable to download API page'): | |
108 | if not self._WORKING_APP_VERSION: | |
109 | app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0] | |
110 | manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0] | |
111 | if app_version and manifest_app_version: | |
112 | self._WORKING_APP_VERSION = (app_version, manifest_app_version) | |
113 | self.write_debug('Imported app version combo from extractor arguments') | |
114 | elif app_version or manifest_app_version: | |
115 | self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True) | |
116 | ||
117 | if self._WORKING_APP_VERSION: | |
118 | app_version, manifest_app_version = self._WORKING_APP_VERSION | |
119 | real_query = self._build_api_query(query, app_version, manifest_app_version) | |
120 | return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote) | |
121 | ||
122 | for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1): | |
123 | real_query = self._build_api_query(query, app_version, manifest_app_version) | |
124 | try: | |
125 | res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote) | |
126 | self._WORKING_APP_VERSION = (app_version, manifest_app_version) | |
127 | return res | |
128 | except ExtractorError as e: | |
129 | if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0: | |
130 | if count == len(self._APP_VERSIONS): | |
131 | if fatal: | |
132 | raise e | |
133 | else: | |
134 | self.report_warning(str(e.cause or e.msg)) | |
135 | return | |
136 | self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS))) | |
137 | continue | |
138 | raise e | |
139 | ||
140 | def _extract_aweme_app(self, aweme_id): | |
141 | feed_list = self._call_api( | |
142 | 'feed', {'aweme_id': aweme_id}, aweme_id, note='Downloading video feed', | |
143 | errnote='Unable to download video feed').get('aweme_list') or [] | |
144 | aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None) | |
145 | if not aweme_detail: | |
146 | raise ExtractorError('Unable to find video in feed', video_id=aweme_id) | |
147 | return self._parse_aweme_video_app(aweme_detail) | |
148 | ||
149 | def _get_subtitles(self, aweme_detail, aweme_id): | |
150 | # TODO: Extract text positioning info | |
151 | subtitles = {} | |
152 | # aweme/detail endpoint subs | |
153 | captions_info = traverse_obj( | |
154 | aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict) | |
155 | for caption in captions_info: | |
156 | caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False) | |
157 | if not caption_url: | |
158 | continue | |
159 | caption_json = self._download_json( | |
160 | caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False) | |
161 | if not caption_json: | |
162 | continue | |
163 | subtitles.setdefault(caption.get('language', 'en'), []).append({ | |
164 | 'ext': 'srt', | |
165 | 'data': '\n\n'.join( | |
166 | f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}' | |
167 | for i, line in enumerate(caption_json['utterances']) if line.get('text')) | |
168 | }) | |
169 | # feed endpoint subs | |
170 | if not subtitles: | |
171 | for caption in traverse_obj(aweme_detail, ('video', 'cla_info', 'caption_infos', ...), expected_type=dict): | |
172 | if not caption.get('url'): | |
173 | continue | |
174 | subtitles.setdefault(caption.get('lang') or 'en', []).append({ | |
175 | 'ext': remove_start(caption.get('caption_format'), 'web'), | |
176 | 'url': caption['url'], | |
177 | }) | |
178 | # webpage subs | |
179 | if not subtitles: | |
180 | for caption in traverse_obj(aweme_detail, ('video', 'subtitleInfos', ...), expected_type=dict): | |
181 | if not caption.get('Url'): | |
182 | continue | |
183 | subtitles.setdefault(caption.get('LanguageCodeName') or 'en', []).append({ | |
184 | 'ext': remove_start(caption.get('Format'), 'web'), | |
185 | 'url': caption['Url'], | |
186 | }) | |
187 | return subtitles | |
188 | ||
189 | def _parse_aweme_video_app(self, aweme_detail): | |
190 | aweme_id = aweme_detail['aweme_id'] | |
191 | video_info = aweme_detail['video'] | |
192 | ||
193 | def parse_url_key(url_key): | |
194 | format_id, codec, res, bitrate = self._search_regex( | |
195 | r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key, | |
196 | 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate')) | |
197 | if not format_id: | |
198 | return {}, None | |
199 | return { | |
200 | 'format_id': format_id, | |
201 | 'vcodec': 'h265' if codec == 'bytevc1' else codec, | |
202 | 'tbr': int_or_none(bitrate, scale=1000) or None, | |
203 | 'quality': qualities(self.QUALITIES)(res), | |
204 | }, res | |
205 | ||
206 | known_resolutions = {} | |
207 | ||
208 | def audio_meta(url): | |
209 | ext = determine_ext(url, default_ext='m4a') | |
210 | return { | |
211 | 'format_note': 'Music track', | |
212 | 'ext': ext, | |
213 | 'acodec': 'aac' if ext == 'm4a' else ext, | |
214 | 'vcodec': 'none', | |
215 | 'width': None, | |
216 | 'height': None, | |
217 | } if ext == 'mp3' or '-music-' in url else {} | |
218 | ||
219 | def extract_addr(addr, add_meta={}): | |
220 | parsed_meta, res = parse_url_key(addr.get('url_key', '')) | |
221 | if res: | |
222 | known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height') or addr.get('height')) | |
223 | known_resolutions[res].setdefault('width', add_meta.get('width') or addr.get('width')) | |
224 | parsed_meta.update(known_resolutions.get(res, {})) | |
225 | add_meta.setdefault('height', int_or_none(res[:-1])) | |
226 | return [{ | |
227 | 'url': url, | |
228 | 'filesize': int_or_none(addr.get('data_size')), | |
229 | 'ext': 'mp4', | |
230 | 'acodec': 'aac', | |
231 | 'source_preference': -2 if 'aweme/v1' in url else -1, # Downloads from API might get blocked | |
232 | **add_meta, **parsed_meta, | |
233 | 'format_note': join_nonempty( | |
234 | add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' '), | |
235 | **audio_meta(url), | |
236 | } for url in addr.get('url_list') or []] | |
237 | ||
238 | # Hack: Add direct video links first to prioritize them when removing duplicate formats | |
239 | formats = [] | |
240 | if video_info.get('play_addr'): | |
241 | formats.extend(extract_addr(video_info['play_addr'], { | |
242 | 'format_id': 'play_addr', | |
243 | 'format_note': 'Direct video', | |
244 | 'vcodec': 'h265' if traverse_obj( | |
245 | video_info, 'is_bytevc1', 'is_h265') else 'h264', # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002 | |
246 | 'width': video_info.get('width'), | |
247 | 'height': video_info.get('height'), | |
248 | })) | |
249 | if video_info.get('download_addr'): | |
250 | formats.extend(extract_addr(video_info['download_addr'], { | |
251 | 'format_id': 'download_addr', | |
252 | 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''), | |
253 | 'vcodec': 'h264', | |
254 | 'width': video_info.get('width'), | |
255 | 'height': video_info.get('height'), | |
256 | 'preference': -2 if video_info.get('has_watermark') else -1, | |
257 | })) | |
258 | if video_info.get('play_addr_h264'): | |
259 | formats.extend(extract_addr(video_info['play_addr_h264'], { | |
260 | 'format_id': 'play_addr_h264', | |
261 | 'format_note': 'Direct video', | |
262 | 'vcodec': 'h264', | |
263 | })) | |
264 | if video_info.get('play_addr_bytevc1'): | |
265 | formats.extend(extract_addr(video_info['play_addr_bytevc1'], { | |
266 | 'format_id': 'play_addr_bytevc1', | |
267 | 'format_note': 'Direct video', | |
268 | 'vcodec': 'h265', | |
269 | })) | |
270 | ||
271 | for bitrate in video_info.get('bit_rate', []): | |
272 | if bitrate.get('play_addr'): | |
273 | formats.extend(extract_addr(bitrate['play_addr'], { | |
274 | 'format_id': bitrate.get('gear_name'), | |
275 | 'format_note': 'Playback video', | |
276 | 'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000), | |
277 | 'vcodec': 'h265' if traverse_obj( | |
278 | bitrate, 'is_bytevc1', 'is_h265') else 'h264', | |
279 | 'fps': bitrate.get('FPS'), | |
280 | })) | |
281 | ||
282 | self._remove_duplicate_formats(formats) | |
283 | auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt') | |
284 | if auth_cookie: | |
285 | for f in formats: | |
286 | self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value) | |
287 | ||
288 | thumbnails = [] | |
289 | for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak', | |
290 | 'origin_cover', 'dynamic_cover'): | |
291 | for cover_url in traverse_obj(video_info, (cover_id, 'url_list', ...)): | |
292 | thumbnails.append({ | |
293 | 'id': cover_id, | |
294 | 'url': cover_url, | |
295 | }) | |
296 | ||
297 | stats_info = aweme_detail.get('statistics') or {} | |
298 | author_info = aweme_detail.get('author') or {} | |
299 | music_info = aweme_detail.get('music') or {} | |
300 | user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info, | |
301 | 'sec_uid', 'id', 'uid', 'unique_id', | |
302 | expected_type=str_or_none, get_all=False)) | |
303 | labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str) | |
304 | ||
305 | contained_music_track = traverse_obj( | |
306 | music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str) | |
307 | contained_music_author = traverse_obj( | |
308 | music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str) | |
309 | ||
310 | is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle') | |
311 | if is_generic_og_trackname: | |
312 | music_track, music_author = contained_music_track or 'original sound', contained_music_author | |
313 | else: | |
314 | music_track, music_author = music_info.get('title'), music_info.get('author') | |
315 | ||
316 | return { | |
317 | 'id': aweme_id, | |
318 | 'extractor_key': TikTokIE.ie_key(), | |
319 | 'extractor': TikTokIE.IE_NAME, | |
320 | 'webpage_url': self._create_url(author_info.get('uid'), aweme_id), | |
321 | **traverse_obj(aweme_detail, { | |
322 | 'title': ('desc', {str}), | |
323 | 'description': ('desc', {str}), | |
324 | 'timestamp': ('create_time', {int_or_none}), | |
325 | }), | |
326 | **traverse_obj(stats_info, { | |
327 | 'view_count': 'play_count', | |
328 | 'like_count': 'digg_count', | |
329 | 'repost_count': 'share_count', | |
330 | 'comment_count': 'comment_count', | |
331 | }, expected_type=int_or_none), | |
332 | **traverse_obj(author_info, { | |
333 | 'uploader': 'unique_id', | |
334 | 'uploader_id': 'uid', | |
335 | 'creator': 'nickname', | |
336 | 'channel_id': 'sec_uid', | |
337 | }, expected_type=str_or_none), | |
338 | 'uploader_url': user_url, | |
339 | 'track': music_track, | |
340 | 'album': str_or_none(music_info.get('album')) or None, | |
341 | 'artist': music_author or None, | |
342 | 'formats': formats, | |
343 | 'subtitles': self.extract_subtitles(aweme_detail, aweme_id), | |
344 | 'thumbnails': thumbnails, | |
345 | 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000), | |
346 | 'availability': self._availability( | |
347 | is_private='Private' in labels, | |
348 | needs_subscription='Friends only' in labels, | |
349 | is_unlisted='Followers only' in labels), | |
350 | '_format_sort_fields': ('quality', 'codec', 'size', 'br'), | |
351 | } | |
352 | ||
353 | def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id): | |
354 | video_info = aweme_detail['video'] | |
355 | author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={}) | |
356 | music_info = aweme_detail.get('music') or {} | |
357 | stats_info = aweme_detail.get('stats') or {} | |
358 | channel_id = traverse_obj(author_info or aweme_detail, (('authorSecId', 'secUid'), {str}), get_all=False) | |
359 | user_url = self._UPLOADER_URL_FORMAT % channel_id if channel_id else None | |
360 | ||
361 | formats = [] | |
362 | width = int_or_none(video_info.get('width')) | |
363 | height = int_or_none(video_info.get('height')) | |
364 | ||
365 | for play_url in traverse_obj(video_info, ('playAddr', ((..., 'src'), None), {url_or_none})): | |
366 | formats.append({ | |
367 | 'url': self._proto_relative_url(play_url), | |
368 | 'ext': 'mp4', | |
369 | 'width': width, | |
370 | 'height': height, | |
371 | }) | |
372 | ||
373 | for download_url in traverse_obj(video_info, (('downloadAddr', ('download', 'url')), {url_or_none})): | |
374 | formats.append({ | |
375 | 'format_id': 'download', | |
376 | 'url': self._proto_relative_url(download_url), | |
377 | 'ext': 'mp4', | |
378 | 'width': width, | |
379 | 'height': height, | |
380 | }) | |
381 | ||
382 | self._remove_duplicate_formats(formats) | |
383 | ||
384 | thumbnails = [] | |
385 | for thumb_url in traverse_obj(aweme_detail, ( | |
386 | (None, 'video'), ('thumbnail', 'cover', 'dynamicCover', 'originCover'), {url_or_none})): | |
387 | thumbnails.append({ | |
388 | 'url': self._proto_relative_url(thumb_url), | |
389 | 'width': width, | |
390 | 'height': height, | |
391 | }) | |
392 | ||
393 | return { | |
394 | 'id': video_id, | |
395 | **traverse_obj(aweme_detail, { | |
396 | 'title': ('desc', {str}), | |
397 | 'description': ('desc', {str}), | |
398 | 'duration': ('video', 'duration', {int_or_none}), | |
399 | 'timestamp': ('createTime', {int_or_none}), | |
400 | }), | |
401 | **traverse_obj(author_info or aweme_detail, { | |
402 | 'creator': ('nickname', {str}), | |
403 | 'uploader': (('uniqueId', 'author'), {str}), | |
404 | 'uploader_id': (('authorId', 'uid', 'id'), {str_or_none}), | |
405 | }, get_all=False), | |
406 | **traverse_obj(stats_info, { | |
407 | 'view_count': 'playCount', | |
408 | 'like_count': 'diggCount', | |
409 | 'repost_count': 'shareCount', | |
410 | 'comment_count': 'commentCount', | |
411 | }, expected_type=int_or_none), | |
412 | **traverse_obj(music_info, { | |
413 | 'track': 'title', | |
414 | 'album': ('album', {lambda x: x or None}), | |
415 | 'artist': 'authorName', | |
416 | }, expected_type=str), | |
417 | 'channel_id': channel_id, | |
418 | 'uploader_url': user_url, | |
419 | 'formats': formats, | |
420 | 'thumbnails': thumbnails, | |
421 | 'http_headers': { | |
422 | 'Referer': webpage_url, | |
423 | } | |
424 | } | |
425 | ||
426 | ||
427 | class TikTokIE(TikTokBaseIE): | |
428 | _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)?/video)/(?P<id>\d+)' | |
429 | _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})'] | |
430 | ||
431 | _TESTS = [{ | |
432 | 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610', | |
433 | 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7', | |
434 | 'info_dict': { | |
435 | 'id': '6748451240264420610', | |
436 | 'ext': 'mp4', | |
437 | 'title': '#jassmanak #lehanga #leenabhushan', | |
438 | 'description': '#jassmanak #lehanga #leenabhushan', | |
439 | 'duration': 13, | |
440 | 'height': 1024, | |
441 | 'width': 576, | |
442 | 'uploader': 'leenabhushan', | |
443 | 'uploader_id': '6691488002098119685', | |
444 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy', | |
445 | 'creator': 'facestoriesbyleenabh', | |
446 | 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', | |
447 | 'upload_date': '20191016', | |
448 | 'timestamp': 1571246252, | |
449 | 'view_count': int, | |
450 | 'like_count': int, | |
451 | 'repost_count': int, | |
452 | 'comment_count': int, | |
453 | 'artist': 'Ysrbeats', | |
454 | 'album': 'Lehanga', | |
455 | 'track': 'Lehanga', | |
456 | }, | |
457 | 'skip': '404 Not Found', | |
458 | }, { | |
459 | 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en', | |
460 | 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b', | |
461 | 'info_dict': { | |
462 | 'id': '6742501081818877190', | |
463 | 'ext': 'mp4', | |
464 | 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94', | |
465 | 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94', | |
466 | 'duration': 27, | |
467 | 'height': 960, | |
468 | 'width': 540, | |
469 | 'uploader': 'patrox', | |
470 | 'uploader_id': '18702747', | |
471 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws', | |
472 | 'channel_id': 'MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws', | |
473 | 'creator': 'patroX', | |
474 | 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', | |
475 | 'upload_date': '20190930', | |
476 | 'timestamp': 1569860870, | |
477 | 'view_count': int, | |
478 | 'like_count': int, | |
479 | 'repost_count': int, | |
480 | 'comment_count': int, | |
481 | 'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson', | |
482 | 'track': 'Big Fun', | |
483 | }, | |
484 | }, { | |
485 | # Banned audio, only available on the app | |
486 | 'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402', | |
487 | 'info_dict': { | |
488 | 'id': '6984138651336838402', | |
489 | 'ext': 'mp4', | |
490 | 'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥', | |
491 | 'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥', | |
492 | 'uploader': 'barudakhb_', | |
493 | 'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6', | |
494 | 'uploader_id': '6974687867511718913', | |
495 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d', | |
496 | 'channel_id': 'MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d', | |
497 | 'track': 'Boka Dance', | |
498 | 'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6', | |
499 | 'timestamp': 1626121503, | |
500 | 'duration': 18, | |
501 | 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', | |
502 | 'upload_date': '20210712', | |
503 | 'view_count': int, | |
504 | 'like_count': int, | |
505 | 'repost_count': int, | |
506 | 'comment_count': int, | |
507 | }, | |
508 | }, { | |
509 | # Sponsored video, only available with feed workaround | |
510 | 'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561', | |
511 | 'info_dict': { | |
512 | 'id': '7042692929109986561', | |
513 | 'ext': 'mp4', | |
514 | 'title': 'Slap and Run!', | |
515 | 'description': 'Slap and Run!', | |
516 | 'uploader': 'user440922249', | |
517 | 'creator': 'Slap And Run', | |
518 | 'uploader_id': '7036055384943690754', | |
519 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_', | |
520 | 'channel_id': 'MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_', | |
521 | 'track': 'Promoted Music', | |
522 | 'timestamp': 1639754738, | |
523 | 'duration': 30, | |
524 | 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', | |
525 | 'upload_date': '20211217', | |
526 | 'view_count': int, | |
527 | 'like_count': int, | |
528 | 'repost_count': int, | |
529 | 'comment_count': int, | |
530 | }, | |
531 | 'params': {'skip_download': True}, # XXX: unable to download video data: HTTP Error 403: Forbidden | |
532 | }, { | |
533 | # Video without title and description | |
534 | 'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694', | |
535 | 'info_dict': { | |
536 | 'id': '7059698374567611694', | |
537 | 'ext': 'mp4', | |
538 | 'title': 'TikTok video #7059698374567611694', | |
539 | 'description': '', | |
540 | 'uploader': 'pokemonlife22', | |
541 | 'creator': 'Pokemon', | |
542 | 'uploader_id': '6820838815978423302', | |
543 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W', | |
544 | 'channel_id': 'MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W', | |
545 | 'track': 'original sound', | |
546 | 'timestamp': 1643714123, | |
547 | 'duration': 6, | |
548 | 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', | |
549 | 'upload_date': '20220201', | |
550 | 'artist': 'Pokemon', | |
551 | 'view_count': int, | |
552 | 'like_count': int, | |
553 | 'repost_count': int, | |
554 | 'comment_count': int, | |
555 | }, | |
556 | }, { | |
557 | # hydration JSON is sent in a <script> element | |
558 | 'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713', | |
559 | 'info_dict': { | |
560 | 'id': '7065799023130643713', | |
561 | 'ext': 'mp4', | |
562 | 'title': '#denidil#денидил', | |
563 | 'description': '#denidil#денидил', | |
564 | 'uploader': 'denidil6', | |
565 | 'uploader_id': '7046664115636405250', | |
566 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ', | |
567 | 'artist': 'Holocron Music', | |
568 | 'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night', | |
569 | 'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night', | |
570 | 'timestamp': 1645134536, | |
571 | 'duration': 26, | |
572 | 'upload_date': '20220217', | |
573 | 'view_count': int, | |
574 | 'like_count': int, | |
575 | 'repost_count': int, | |
576 | 'comment_count': int, | |
577 | }, | |
578 | 'skip': 'This video is unavailable', | |
579 | }, { | |
580 | # slideshow audio-only mp3 format | |
581 | 'url': 'https://www.tiktok.com/@_le_cannibale_/video/7139980461132074283', | |
582 | 'info_dict': { | |
583 | 'id': '7139980461132074283', | |
584 | 'ext': 'mp3', | |
585 | 'title': 'TikTok video #7139980461132074283', | |
586 | 'description': '', | |
587 | 'creator': 'Antaura', | |
588 | 'uploader': '_le_cannibale_', | |
589 | 'uploader_id': '6604511138619654149', | |
590 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP', | |
591 | 'channel_id': 'MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP', | |
592 | 'artist': 'nathan !', | |
593 | 'track': 'grahamscott canon', | |
594 | 'upload_date': '20220905', | |
595 | 'timestamp': 1662406249, | |
596 | 'view_count': int, | |
597 | 'like_count': int, | |
598 | 'repost_count': int, | |
599 | 'comment_count': int, | |
600 | 'thumbnail': r're:^https://.+\.webp', | |
601 | }, | |
602 | }, { | |
603 | # only available via web | |
604 | 'url': 'https://www.tiktok.com/@moxypatch/video/7206382937372134662', | |
605 | 'md5': '6aba7fad816e8709ff2c149679ace165', | |
606 | 'info_dict': { | |
607 | 'id': '7206382937372134662', | |
608 | 'ext': 'mp4', | |
609 | 'title': 'md5:1d95c0b96560ca0e8a231af4172b2c0a', | |
610 | 'description': 'md5:1d95c0b96560ca0e8a231af4172b2c0a', | |
611 | 'creator': 'MoxyPatch', | |
612 | 'uploader': 'moxypatch', | |
613 | 'uploader_id': '7039142049363379205', | |
614 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V', | |
615 | 'channel_id': 'MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V', | |
616 | 'artist': 'your worst nightmare', | |
617 | 'track': 'original sound', | |
618 | 'upload_date': '20230303', | |
619 | 'timestamp': 1677866781, | |
620 | 'duration': 10, | |
621 | 'view_count': int, | |
622 | 'like_count': int, | |
623 | 'repost_count': int, | |
624 | 'comment_count': int, | |
625 | 'thumbnail': r're:^https://.+', | |
626 | 'thumbnails': 'count:3', | |
627 | }, | |
628 | 'expected_warnings': ['Unable to find video in feed'], | |
629 | }, { | |
630 | # 1080p format | |
631 | 'url': 'https://www.tiktok.com/@tatemcrae/video/7107337212743830830', | |
632 | 'md5': '982512017a8a917124d5a08c8ae79621', | |
633 | 'info_dict': { | |
634 | 'id': '7107337212743830830', | |
635 | 'ext': 'mp4', | |
636 | 'title': 'new music video 4 don’t come backkkk🧸🖤 i hope u enjoy !! @musicontiktok', | |
637 | 'description': 'new music video 4 don’t come backkkk🧸🖤 i hope u enjoy !! @musicontiktok', | |
638 | 'uploader': 'tatemcrae', | |
639 | 'uploader_id': '86328792343818240', | |
640 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd', | |
641 | 'channel_id': 'MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd', | |
642 | 'creator': 'tate mcrae', | |
643 | 'artist': 'tate mcrae', | |
644 | 'track': 'original sound', | |
645 | 'upload_date': '20220609', | |
646 | 'timestamp': 1654805899, | |
647 | 'duration': 150, | |
648 | 'view_count': int, | |
649 | 'like_count': int, | |
650 | 'repost_count': int, | |
651 | 'comment_count': int, | |
652 | 'thumbnail': r're:^https://.+\.webp', | |
653 | }, | |
654 | 'params': {'format': 'bytevc1_1080p_808907-0'}, | |
655 | }, { | |
656 | # Slideshow, audio-only m4a format | |
657 | 'url': 'https://www.tiktok.com/@hara_yoimiya/video/7253412088251534594', | |
658 | 'md5': '2ff8fe0174db2dbf49c597a7bef4e47d', | |
659 | 'info_dict': { | |
660 | 'id': '7253412088251534594', | |
661 | 'ext': 'm4a', | |
662 | 'title': 'я ред флаг простите #переписка #щитпост #тревожныйтиппривязанности #рекомендации ', | |
663 | 'description': 'я ред флаг простите #переписка #щитпост #тревожныйтиппривязанности #рекомендации ', | |
664 | 'uploader': 'hara_yoimiya', | |
665 | 'uploader_id': '6582536342634676230', | |
666 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB', | |
667 | 'channel_id': 'MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB', | |
668 | 'creator': 'лампочка', | |
669 | 'artist': 'Øneheart', | |
670 | 'album': 'watching the stars', | |
671 | 'track': 'watching the stars', | |
672 | 'upload_date': '20230708', | |
673 | 'timestamp': 1688816612, | |
674 | 'view_count': int, | |
675 | 'like_count': int, | |
676 | 'comment_count': int, | |
677 | 'repost_count': int, | |
678 | 'thumbnail': r're:^https://.+\.webp', | |
679 | }, | |
680 | }, { | |
681 | # Auto-captions available | |
682 | 'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758', | |
683 | 'only_matching': True | |
684 | }] | |
685 | ||
686 | def _real_extract(self, url): | |
687 | video_id, user_id = self._match_valid_url(url).group('id', 'user_id') | |
688 | try: | |
689 | return self._extract_aweme_app(video_id) | |
690 | except ExtractorError as e: | |
691 | self.report_warning(f'{e}; trying with webpage') | |
692 | ||
693 | url = self._create_url(user_id, video_id) | |
694 | webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'Mozilla/5.0'}) | |
695 | next_data = self._search_nextjs_data(webpage, video_id, default='{}') | |
696 | if next_data: | |
697 | status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0 | |
698 | video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict) | |
699 | else: | |
700 | sigi_data = self._get_sigi_state(webpage, video_id) | |
701 | status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0 | |
702 | video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict) | |
703 | ||
704 | if status == 0: | |
705 | return self._parse_aweme_video_web(video_data, url, video_id) | |
706 | elif status == 10216: | |
707 | raise ExtractorError('This video is private', expected=True) | |
708 | raise ExtractorError('Video not available', video_id=video_id) | |
709 | ||
710 | ||
711 | class TikTokUserIE(TikTokBaseIE): | |
712 | IE_NAME = 'tiktok:user' | |
713 | _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])' | |
714 | _WORKING = False | |
715 | _TESTS = [{ | |
716 | 'url': 'https://tiktok.com/@corgibobaa?lang=en', | |
717 | 'playlist_mincount': 45, | |
718 | 'info_dict': { | |
719 | 'id': '6935371178089399301', | |
720 | 'title': 'corgibobaa', | |
721 | 'thumbnail': r're:https://.+_1080x1080\.webp' | |
722 | }, | |
723 | 'expected_warnings': ['Retrying'] | |
724 | }, { | |
725 | 'url': 'https://www.tiktok.com/@6820838815978423302', | |
726 | 'playlist_mincount': 5, | |
727 | 'info_dict': { | |
728 | 'id': '6820838815978423302', | |
729 | 'title': '6820838815978423302', | |
730 | 'thumbnail': r're:https://.+_1080x1080\.webp' | |
731 | }, | |
732 | 'expected_warnings': ['Retrying'] | |
733 | }, { | |
734 | 'url': 'https://www.tiktok.com/@meme', | |
735 | 'playlist_mincount': 593, | |
736 | 'info_dict': { | |
737 | 'id': '79005827461758976', | |
738 | 'title': 'meme', | |
739 | 'thumbnail': r're:https://.+_1080x1080\.webp' | |
740 | }, | |
741 | 'expected_warnings': ['Retrying'] | |
742 | }] | |
743 | ||
744 | r''' # TODO: Fix by adding _signature to api_url | |
745 | def _entries(self, webpage, user_id, username): | |
746 | secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username) | |
747 | verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id') | |
748 | if not verifyfp_cookie: | |
749 | raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True) | |
750 | api_url = f'https://m.tiktok.com/api/post/item_list/?aid=1988&cookie_enabled=true&count=30&verifyFp={verifyfp_cookie.value}&secUid={secuid}&cursor=' | |
751 | cursor = '0' | |
752 | for page in itertools.count(): | |
753 | data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page) | |
754 | for video in data_json.get('itemList', []): | |
755 | video_id = video['id'] | |
756 | video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}' | |
757 | yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc'))) | |
758 | if not data_json.get('hasMore'): | |
759 | break | |
760 | cursor = data_json['cursor'] | |
761 | ''' | |
762 | ||
763 | def _video_entries_api(self, webpage, user_id, username): | |
764 | query = { | |
765 | 'user_id': user_id, | |
766 | 'count': 21, | |
767 | 'max_cursor': 0, | |
768 | 'min_cursor': 0, | |
769 | 'retry_type': 'no_retry', | |
770 | 'device_id': ''.join(random.choices(string.digits, k=19)), # Some endpoints don't like randomized device_id, so it isn't directly set in _call_api. | |
771 | } | |
772 | ||
773 | for page in itertools.count(1): | |
774 | for retry in self.RetryManager(): | |
775 | try: | |
776 | post_list = self._call_api( | |
777 | 'aweme/post', query, username, note=f'Downloading user video list page {page}', | |
778 | errnote='Unable to download user video list') | |
779 | except ExtractorError as e: | |
780 | if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0: | |
781 | retry.error = e | |
782 | continue | |
783 | raise | |
784 | yield from post_list.get('aweme_list', []) | |
785 | if not post_list.get('has_more'): | |
786 | break | |
787 | query['max_cursor'] = post_list['max_cursor'] | |
788 | ||
789 | def _entries_api(self, user_id, videos): | |
790 | for video in videos: | |
791 | yield { | |
792 | **self._parse_aweme_video_app(video), | |
793 | 'extractor_key': TikTokIE.ie_key(), | |
794 | 'extractor': 'TikTok', | |
795 | 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}', | |
796 | } | |
797 | ||
798 | def _real_extract(self, url): | |
799 | user_name = self._match_id(url) | |
800 | webpage = self._download_webpage(url, user_name, headers={ | |
801 | 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)' | |
802 | }) | |
803 | user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name | |
804 | ||
805 | videos = LazyList(self._video_entries_api(webpage, user_id, user_name)) | |
806 | thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0)) | |
807 | ||
808 | return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail) | |
809 | ||
810 | ||
811 | class TikTokBaseListIE(TikTokBaseIE): # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor | |
812 | def _entries(self, list_id, display_id): | |
813 | query = { | |
814 | self._QUERY_NAME: list_id, | |
815 | 'cursor': 0, | |
816 | 'count': 20, | |
817 | 'type': 5, | |
818 | 'device_id': ''.join(random.choices(string.digits, k=19)) | |
819 | } | |
820 | ||
821 | for page in itertools.count(1): | |
822 | for retry in self.RetryManager(): | |
823 | try: | |
824 | post_list = self._call_api( | |
825 | self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}', | |
826 | errnote='Unable to download video list') | |
827 | except ExtractorError as e: | |
828 | if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0: | |
829 | retry.error = e | |
830 | continue | |
831 | raise | |
832 | for video in post_list.get('aweme_list', []): | |
833 | yield { | |
834 | **self._parse_aweme_video_app(video), | |
835 | 'extractor_key': TikTokIE.ie_key(), | |
836 | 'extractor': 'TikTok', | |
837 | 'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}', | |
838 | } | |
839 | if not post_list.get('has_more'): | |
840 | break | |
841 | query['cursor'] = post_list['cursor'] | |
842 | ||
843 | def _real_extract(self, url): | |
844 | list_id = self._match_id(url) | |
845 | return self.playlist_result(self._entries(list_id, list_id), list_id) | |
846 | ||
847 | ||
848 | class TikTokSoundIE(TikTokBaseListIE): | |
849 | IE_NAME = 'tiktok:sound' | |
850 | _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?' | |
851 | _WORKING = False | |
852 | _QUERY_NAME = 'music_id' | |
853 | _API_ENDPOINT = 'music/aweme' | |
854 | _TESTS = [{ | |
855 | 'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en', | |
856 | 'playlist_mincount': 100, | |
857 | 'info_dict': { | |
858 | 'id': '6956990112127585029' | |
859 | }, | |
860 | 'expected_warnings': ['Retrying'] | |
861 | }, { | |
862 | # Actual entries are less than listed video count | |
863 | 'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381', | |
864 | 'playlist_mincount': 2182, | |
865 | 'info_dict': { | |
866 | 'id': '7036843036118469381' | |
867 | }, | |
868 | 'expected_warnings': ['Retrying'] | |
869 | }] | |
870 | ||
871 | ||
872 | class TikTokEffectIE(TikTokBaseListIE): | |
873 | IE_NAME = 'tiktok:effect' | |
874 | _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?' | |
875 | _WORKING = False | |
876 | _QUERY_NAME = 'sticker_id' | |
877 | _API_ENDPOINT = 'sticker/aweme' | |
878 | _TESTS = [{ | |
879 | 'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156', | |
880 | 'playlist_mincount': 100, | |
881 | 'info_dict': { | |
882 | 'id': '1258156', | |
883 | }, | |
884 | 'expected_warnings': ['Retrying'] | |
885 | }, { | |
886 | # Different entries between mobile and web, depending on region | |
887 | 'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565', | |
888 | 'only_matching': True | |
889 | }] | |
890 | ||
891 | ||
892 | class TikTokTagIE(TikTokBaseListIE): | |
893 | IE_NAME = 'tiktok:tag' | |
894 | _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)' | |
895 | _WORKING = False | |
896 | _QUERY_NAME = 'ch_id' | |
897 | _API_ENDPOINT = 'challenge/aweme' | |
898 | _TESTS = [{ | |
899 | 'url': 'https://tiktok.com/tag/hello2018', | |
900 | 'playlist_mincount': 39, | |
901 | 'info_dict': { | |
902 | 'id': '46294678', | |
903 | 'title': 'hello2018', | |
904 | }, | |
905 | 'expected_warnings': ['Retrying'] | |
906 | }, { | |
907 | 'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1', | |
908 | 'only_matching': True | |
909 | }] | |
910 | ||
911 | def _real_extract(self, url): | |
912 | display_id = self._match_id(url) | |
913 | webpage = self._download_webpage(url, display_id, headers={ | |
914 | 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)' | |
915 | }) | |
916 | tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID') | |
917 | return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id) | |
918 | ||
919 | ||
920 | class DouyinIE(TikTokBaseIE): | |
921 | _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)' | |
922 | _TESTS = [{ | |
923 | 'url': 'https://www.douyin.com/video/6961737553342991651', | |
924 | 'md5': 'a97db7e3e67eb57bf40735c022ffa228', | |
925 | 'info_dict': { | |
926 | 'id': '6961737553342991651', | |
927 | 'ext': 'mp4', | |
928 | 'title': '#杨超越 小小水手带你去远航❤️', | |
929 | 'description': '#杨超越 小小水手带你去远航❤️', | |
930 | 'uploader_id': '110403406559', | |
931 | 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', | |
932 | 'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', | |
933 | 'creator': '杨超越', | |
934 | 'duration': 19782, | |
935 | 'timestamp': 1620905839, | |
936 | 'upload_date': '20210513', | |
937 | 'track': '@杨超越创作的原声', | |
938 | 'view_count': int, | |
939 | 'like_count': int, | |
940 | 'repost_count': int, | |
941 | 'comment_count': int, | |
942 | 'thumbnail': r're:https?://.+\.jpe?g', | |
943 | }, | |
944 | }, { | |
945 | 'url': 'https://www.douyin.com/video/6982497745948921092', | |
946 | 'md5': '34a87ebff3833357733da3fe17e37c0e', | |
947 | 'info_dict': { | |
948 | 'id': '6982497745948921092', | |
949 | 'ext': 'mp4', | |
950 | 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想', | |
951 | 'description': '这个夏日和小羊@杨超越 一起遇见白色幻想', | |
952 | 'uploader_id': '408654318141572', | |
953 | 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA', | |
954 | 'channel_id': 'MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA', | |
955 | 'creator': '杨超越工作室', | |
956 | 'duration': 42479, | |
957 | 'timestamp': 1625739481, | |
958 | 'upload_date': '20210708', | |
959 | 'track': '@杨超越工作室创作的原声', | |
960 | 'view_count': int, | |
961 | 'like_count': int, | |
962 | 'repost_count': int, | |
963 | 'comment_count': int, | |
964 | 'thumbnail': r're:https?://.+\.jpe?g', | |
965 | }, | |
966 | }, { | |
967 | 'url': 'https://www.douyin.com/video/6953975910773099811', | |
968 | 'md5': 'dde3302460f19db59c47060ff013b902', | |
969 | 'info_dict': { | |
970 | 'id': '6953975910773099811', | |
971 | 'ext': 'mp4', | |
972 | 'title': '#一起看海 出现在你的夏日里', | |
973 | 'description': '#一起看海 出现在你的夏日里', | |
974 | 'uploader_id': '110403406559', | |
975 | 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', | |
976 | 'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', | |
977 | 'creator': '杨超越', | |
978 | 'duration': 17343, | |
979 | 'timestamp': 1619098692, | |
980 | 'upload_date': '20210422', | |
981 | 'track': '@杨超越创作的原声', | |
982 | 'view_count': int, | |
983 | 'like_count': int, | |
984 | 'repost_count': int, | |
985 | 'comment_count': int, | |
986 | 'thumbnail': r're:https?://.+\.jpe?g', | |
987 | }, | |
988 | }, { | |
989 | 'url': 'https://www.douyin.com/video/6950251282489675042', | |
990 | 'md5': 'b4db86aec367ef810ddd38b1737d2fed', | |
991 | 'info_dict': { | |
992 | 'id': '6950251282489675042', | |
993 | 'ext': 'mp4', | |
994 | 'title': '哈哈哈,成功了哈哈哈哈哈哈', | |
995 | 'uploader': '杨超越', | |
996 | 'upload_date': '20210412', | |
997 | 'timestamp': 1618231483, | |
998 | 'uploader_id': '110403406559', | |
999 | 'view_count': int, | |
1000 | 'like_count': int, | |
1001 | 'repost_count': int, | |
1002 | 'comment_count': int, | |
1003 | }, | |
1004 | 'skip': 'No longer available', | |
1005 | }, { | |
1006 | 'url': 'https://www.douyin.com/video/6963263655114722595', | |
1007 | 'md5': 'cf9f11f0ec45d131445ec2f06766e122', | |
1008 | 'info_dict': { | |
1009 | 'id': '6963263655114722595', | |
1010 | 'ext': 'mp4', | |
1011 | 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈', | |
1012 | 'description': '#哪个爱豆的105度最甜 换个角度看看我哈哈', | |
1013 | 'uploader_id': '110403406559', | |
1014 | 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', | |
1015 | 'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', | |
1016 | 'creator': '杨超越', | |
1017 | 'duration': 15115, | |
1018 | 'timestamp': 1621261163, | |
1019 | 'upload_date': '20210517', | |
1020 | 'track': '@杨超越创作的原声', | |
1021 | 'view_count': int, | |
1022 | 'like_count': int, | |
1023 | 'repost_count': int, | |
1024 | 'comment_count': int, | |
1025 | 'thumbnail': r're:https?://.+\.jpe?g', | |
1026 | }, | |
1027 | }] | |
1028 | _APP_VERSIONS = [('23.3.0', '230300')] | |
1029 | _APP_NAME = 'aweme' | |
1030 | _AID = 1128 | |
1031 | _API_HOSTNAME = 'aweme.snssdk.com' | |
1032 | _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s' | |
1033 | _WEBPAGE_HOST = 'https://www.douyin.com/' | |
1034 | ||
1035 | def _real_extract(self, url): | |
1036 | video_id = self._match_id(url) | |
1037 | ||
1038 | try: | |
1039 | return self._extract_aweme_app(video_id) | |
1040 | except ExtractorError as e: | |
1041 | e.expected = True | |
1042 | self.to_screen(f'{e}; trying with webpage') | |
1043 | ||
1044 | webpage = self._download_webpage(url, video_id) | |
1045 | render_data = self._search_json( | |
1046 | r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>', webpage, 'render data', video_id, | |
1047 | contains_pattern=r'%7B(?s:.+)%7D', fatal=False, transform_source=compat_urllib_parse_unquote) | |
1048 | if not render_data: | |
1049 | # TODO: Run verification challenge code to generate signature cookies | |
1050 | cookies = self._get_cookies(self._WEBPAGE_HOST) | |
1051 | expected = not cookies.get('s_v_web_id') or not cookies.get('ttwid') | |
1052 | raise ExtractorError( | |
1053 | 'Fresh cookies (not necessarily logged in) are needed', expected=expected) | |
1054 | ||
1055 | return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url, video_id) | |
1056 | ||
1057 | ||
1058 | class TikTokVMIE(InfoExtractor): | |
1059 | _VALID_URL = r'https?://(?:(?:vm|vt)\.tiktok\.com|(?:www\.)tiktok\.com/t)/(?P<id>\w+)' | |
1060 | IE_NAME = 'vm.tiktok' | |
1061 | ||
1062 | _TESTS = [{ | |
1063 | 'url': 'https://www.tiktok.com/t/ZTRC5xgJp', | |
1064 | 'info_dict': { | |
1065 | 'id': '7170520270497680683', | |
1066 | 'ext': 'mp4', | |
1067 | 'title': 'md5:c64f6152330c2efe98093ccc8597871c', | |
1068 | 'uploader_id': '6687535061741700102', | |
1069 | 'upload_date': '20221127', | |
1070 | 'view_count': int, | |
1071 | 'like_count': int, | |
1072 | 'comment_count': int, | |
1073 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAObqu3WCTXxmw2xwZ3iLEHnEecEIw7ks6rxWqOqOhaPja9BI7gqUQnjw8_5FSoDXX', | |
1074 | 'album': 'Wave of Mutilation: Best of Pixies', | |
1075 | 'thumbnail': r're:https://.+\.webp.*', | |
1076 | 'duration': 5, | |
1077 | 'timestamp': 1669516858, | |
1078 | 'repost_count': int, | |
1079 | 'artist': 'Pixies', | |
1080 | 'track': 'Where Is My Mind?', | |
1081 | 'description': 'md5:c64f6152330c2efe98093ccc8597871c', | |
1082 | 'uploader': 'sigmachaddeus', | |
1083 | 'creator': 'SigmaChad', | |
1084 | }, | |
1085 | }, { | |
1086 | 'url': 'https://vm.tiktok.com/ZTR45GpSF/', | |
1087 | 'info_dict': { | |
1088 | 'id': '7106798200794926362', | |
1089 | 'ext': 'mp4', | |
1090 | 'title': 'md5:edc3e7ea587847f8537468f2fe51d074', | |
1091 | 'uploader_id': '6997695878846268418', | |
1092 | 'upload_date': '20220608', | |
1093 | 'view_count': int, | |
1094 | 'like_count': int, | |
1095 | 'comment_count': int, | |
1096 | 'thumbnail': r're:https://.+\.webp.*', | |
1097 | 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAdZ_NcPPgMneaGrW0hN8O_J_bwLshwNNERRF5DxOw2HKIzk0kdlLrR8RkVl1ksrMO', | |
1098 | 'duration': 29, | |
1099 | 'timestamp': 1654680400, | |
1100 | 'repost_count': int, | |
1101 | 'artist': 'Akihitoko', | |
1102 | 'track': 'original sound', | |
1103 | 'description': 'md5:edc3e7ea587847f8537468f2fe51d074', | |
1104 | 'uploader': 'akihitoko1', | |
1105 | 'creator': 'Akihitoko', | |
1106 | }, | |
1107 | }, { | |
1108 | 'url': 'https://vt.tiktok.com/ZSe4FqkKd', | |
1109 | 'only_matching': True, | |
1110 | }] | |
1111 | ||
1112 | def _real_extract(self, url): | |
1113 | new_url = self._request_webpage( | |
1114 | HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).url | |
1115 | if self.suitable(new_url): # Prevent infinite loop in case redirect fails | |
1116 | raise UnsupportedError(new_url) | |
1117 | return self.url_result(new_url) | |
1118 | ||
1119 | ||
1120 | class TikTokLiveIE(TikTokBaseIE): | |
1121 | _VALID_URL = r'''(?x)https?://(?: | |
1122 | (?:www\.)?tiktok\.com/@(?P<uploader>[\w.-]+)/live| | |
1123 | m\.tiktok\.com/share/live/(?P<id>\d+) | |
1124 | )''' | |
1125 | IE_NAME = 'tiktok:live' | |
1126 | ||
1127 | _TESTS = [{ | |
1128 | 'url': 'https://www.tiktok.com/@weathernewslive/live', | |
1129 | 'info_dict': { | |
1130 | 'id': '7210809319192726273', | |
1131 | 'ext': 'mp4', | |
1132 | 'title': r're:ウェザーニュースLiVE[\d\s:-]*', | |
1133 | 'creator': 'ウェザーニュースLiVE', | |
1134 | 'uploader': 'weathernewslive', | |
1135 | 'uploader_id': '6621496731283095554', | |
1136 | 'uploader_url': 'https://www.tiktok.com/@weathernewslive', | |
1137 | 'live_status': 'is_live', | |
1138 | 'concurrent_view_count': int, | |
1139 | }, | |
1140 | 'params': {'skip_download': 'm3u8'}, | |
1141 | }, { | |
1142 | 'url': 'https://www.tiktok.com/@pilarmagenta/live', | |
1143 | 'info_dict': { | |
1144 | 'id': '7209423610325322522', | |
1145 | 'ext': 'mp4', | |
1146 | 'title': str, | |
1147 | 'creator': 'Pilarmagenta', | |
1148 | 'uploader': 'pilarmagenta', | |
1149 | 'uploader_id': '6624846890674683909', | |
1150 | 'uploader_url': 'https://www.tiktok.com/@pilarmagenta', | |
1151 | 'live_status': 'is_live', | |
1152 | 'concurrent_view_count': int, | |
1153 | }, | |
1154 | 'skip': 'Livestream', | |
1155 | }, { | |
1156 | 'url': 'https://m.tiktok.com/share/live/7209423610325322522/?language=en', | |
1157 | 'only_matching': True, | |
1158 | }, { | |
1159 | 'url': 'https://www.tiktok.com/@iris04201/live', | |
1160 | 'only_matching': True, | |
1161 | }] | |
1162 | ||
1163 | def _call_api(self, url, param, room_id, uploader, key=None): | |
1164 | response = traverse_obj(self._download_json( | |
1165 | url, room_id, fatal=False, query={ | |
1166 | 'aid': '1988', | |
1167 | param: room_id, | |
1168 | }), (key, {dict}), default={}) | |
1169 | ||
1170 | # status == 2 if live else 4 | |
1171 | if int_or_none(response.get('status')) == 2: | |
1172 | return response | |
1173 | # If room_id is obtained via mobile share URL and cannot be refreshed, do not wait for live | |
1174 | elif not uploader: | |
1175 | raise ExtractorError('This livestream has ended', expected=True) | |
1176 | raise UserNotLive(video_id=uploader) | |
1177 | ||
1178 | def _real_extract(self, url): | |
1179 | uploader, room_id = self._match_valid_url(url).group('uploader', 'id') | |
1180 | webpage = self._download_webpage( | |
1181 | url, uploader or room_id, headers={'User-Agent': 'Mozilla/5.0'}, fatal=not room_id) | |
1182 | ||
1183 | if webpage: | |
1184 | data = try_call(lambda: self._get_sigi_state(webpage, uploader or room_id)) | |
1185 | room_id = (traverse_obj(data, ('UserModule', 'users', ..., 'roomId', {str_or_none}), get_all=False) | |
1186 | or self._search_regex(r'snssdk\d*://live\?room_id=(\d+)', webpage, 'room ID', default=None) | |
1187 | or room_id) | |
1188 | uploader = uploader or traverse_obj( | |
1189 | data, ('LiveRoom', 'liveRoomUserInfo', 'user', 'uniqueId'), | |
1190 | ('UserModule', 'users', ..., 'uniqueId'), get_all=False, expected_type=str) | |
1191 | ||
1192 | if not room_id: | |
1193 | raise UserNotLive(video_id=uploader) | |
1194 | ||
1195 | formats = [] | |
1196 | live_info = self._call_api( | |
1197 | 'https://webcast.tiktok.com/webcast/room/info', 'room_id', room_id, uploader, key='data') | |
1198 | ||
1199 | get_quality = qualities(('SD1', 'ld', 'SD2', 'sd', 'HD1', 'hd', 'FULL_HD1', 'uhd', 'ORIGION', 'origin')) | |
1200 | parse_inner = lambda x: self._parse_json(x, None) | |
1201 | ||
1202 | for quality, stream in traverse_obj(live_info, ( | |
1203 | 'stream_url', 'live_core_sdk_data', 'pull_data', 'stream_data', | |
1204 | {parse_inner}, 'data', {dict}), default={}).items(): | |
1205 | ||
1206 | sdk_params = traverse_obj(stream, ('main', 'sdk_params', {parse_inner}, { | |
1207 | 'vcodec': ('VCodec', {str}), | |
1208 | 'tbr': ('vbitrate', {lambda x: int_or_none(x, 1000)}), | |
1209 | 'resolution': ('resolution', {lambda x: re.match(r'(?i)\d+x\d+|\d+p', x).group().lower()}), | |
1210 | })) | |
1211 | ||
1212 | flv_url = traverse_obj(stream, ('main', 'flv', {url_or_none})) | |
1213 | if flv_url: | |
1214 | formats.append({ | |
1215 | 'url': flv_url, | |
1216 | 'ext': 'flv', | |
1217 | 'format_id': f'flv-{quality}', | |
1218 | 'quality': get_quality(quality), | |
1219 | **sdk_params, | |
1220 | }) | |
1221 | ||
1222 | hls_url = traverse_obj(stream, ('main', 'hls', {url_or_none})) | |
1223 | if hls_url: | |
1224 | formats.append({ | |
1225 | 'url': hls_url, | |
1226 | 'ext': 'mp4', | |
1227 | 'protocol': 'm3u8_native', | |
1228 | 'format_id': f'hls-{quality}', | |
1229 | 'quality': get_quality(quality), | |
1230 | **sdk_params, | |
1231 | }) | |
1232 | ||
1233 | def get_vcodec(*keys): | |
1234 | return traverse_obj(live_info, ( | |
1235 | 'stream_url', *keys, {parse_inner}, 'VCodec', {str})) | |
1236 | ||
1237 | for stream in ('hls', 'rtmp'): | |
1238 | stream_url = traverse_obj(live_info, ('stream_url', f'{stream}_pull_url', {url_or_none})) | |
1239 | if stream_url: | |
1240 | formats.append({ | |
1241 | 'url': stream_url, | |
1242 | 'ext': 'mp4' if stream == 'hls' else 'flv', | |
1243 | 'protocol': 'm3u8_native' if stream == 'hls' else 'https', | |
1244 | 'format_id': f'{stream}-pull', | |
1245 | 'vcodec': get_vcodec(f'{stream}_pull_url_params'), | |
1246 | 'quality': get_quality('ORIGION'), | |
1247 | }) | |
1248 | ||
1249 | for f_id, f_url in traverse_obj(live_info, ('stream_url', 'flv_pull_url', {dict}), default={}).items(): | |
1250 | if not url_or_none(f_url): | |
1251 | continue | |
1252 | formats.append({ | |
1253 | 'url': f_url, | |
1254 | 'ext': 'flv', | |
1255 | 'format_id': f'flv-{f_id}'.lower(), | |
1256 | 'vcodec': get_vcodec('flv_pull_url_params', f_id), | |
1257 | 'quality': get_quality(f_id), | |
1258 | }) | |
1259 | ||
1260 | # If uploader is a guest on another's livestream, primary endpoint will not have m3u8 URLs | |
1261 | if not traverse_obj(formats, lambda _, v: v['ext'] == 'mp4'): | |
1262 | live_info = merge_dicts(live_info, self._call_api( | |
1263 | 'https://www.tiktok.com/api/live/detail/', 'roomID', room_id, uploader, key='LiveRoomInfo')) | |
1264 | if url_or_none(live_info.get('liveUrl')): | |
1265 | formats.append({ | |
1266 | 'url': live_info['liveUrl'], | |
1267 | 'ext': 'mp4', | |
1268 | 'protocol': 'm3u8_native', | |
1269 | 'format_id': 'hls-fallback', | |
1270 | 'vcodec': 'h264', | |
1271 | 'quality': get_quality('origin'), | |
1272 | }) | |
1273 | ||
1274 | uploader = uploader or traverse_obj(live_info, ('ownerInfo', 'uniqueId'), ('owner', 'display_id')) | |
1275 | ||
1276 | return { | |
1277 | 'id': room_id, | |
1278 | 'uploader': uploader, | |
1279 | 'uploader_url': format_field(uploader, None, self._UPLOADER_URL_FORMAT) or None, | |
1280 | 'is_live': True, | |
1281 | 'formats': formats, | |
1282 | '_format_sort_fields': ('quality', 'ext'), | |
1283 | **traverse_obj(live_info, { | |
1284 | 'title': 'title', | |
1285 | 'uploader_id': (('ownerInfo', 'owner'), 'id', {str_or_none}), | |
1286 | 'creator': (('ownerInfo', 'owner'), 'nickname'), | |
1287 | 'concurrent_view_count': (('user_count', ('liveRoomStats', 'userCount')), {int_or_none}), | |
1288 | }, get_all=False), | |
1289 | } |